Readmissions are common, costly and often detrimental to patients, yet the relationship of patient, hospital and community factors to hospita readmission rates is unclear. As interest in reducing hospital readmissions rises, it is crucial to understand the type and extent of hospital and community influence on readmission rates so that the best targets for interventions can be identified. Nationally, there is intense focus on reducing readmission rates and in increasing coordination of care, spearheaded by major Federal policy initiatives such as Hospital Engagement Networks, the Community Care Transitions Program, and the Hospital Readmission Reduction Program. Individual hospitals are also engaged in local efforts that maximize area strengths to reduce readmissions. The investigative team has previously developed a new all-condition 30-day risk-standardized hospital-wide unplanned readmission measure. The measure has been endorsed by the National Quality Forum and will be publicly reported by Medicare in 2013. This measure can be used as a validated, risk-standardized assay to compare hospital readmission rates while accounting for differences in diagnoses and case mix among hospitals. Furthermore, the measure is a composite of five specialty-specific sub-measures, allowing us to stratify results by specialty. We will use this assay to examine the influence of hospitals and communities on readmission rates and to study the impact of major policy changes. Specifically, using national Medicare administrative claims data, along with several complementary data sources, our objectives are to: 1) identify characteristics of hospitals and communities associated with readmission rates; 2) determine which hospital and community factors are relevant for each of five clinical specialty cohorts (cardiovascular disease, cardiorespiratory disease, neurological disease, other medical disease, and surgical conditions); and 3) characterize trends in readmission rates between 2007 and 2016, determine the association of improvements with policy changes, and conduct in-depth qualitative analysis of hospitals with greatest and least change over time. This proposal is relevant to several AHRQ priority populations, including low income groups, minority groups, women, older patients, and individuals with special health care needs, and will provide information that advances our understanding of hospital readmissions, enabling clinicians, hospitals, and policymakers to improve healthcare delivery for all patients.